> For the complete documentation index, see [llms.txt](https://docs.darcyiq.com/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://docs.darcyiq.com/build/testdrive/testdrive-with-darcy.md).

# TestDrive with Darcy

Drive TestDrive from chat, from Slack, and from your editor over MCP

TestDrive has a full interface, but most of what it does can be asked for instead — from the workspace chat panel, from a Slack channel, or from your editor.

## Darcy Chat in a Workspace

Open **Darcy Chat** from the workspace header and Darcy arrives already scoped to that workspace. It can look at the tests, the runs, and what the workspace has learned, and it can act on them.

| Ask for           | Example                                                           |
| ----------------- | ----------------------------------------------------------------- |
| **Status**        | "Did last night's smoke run pass?"                                |
| **Investigation** | "Why did the checkout test fail on staging?"                      |
| **New coverage**  | "Add a test that a signed-out visitor can't reach the admin page" |
| **Maintenance**   | "Rebuild the script for the invoice download test"                |
| **A run**         | "Run the billing-tagged tests against staging"                    |

Reading happens directly. Anything that changes the workspace — creating a test, saving a script, starting a run — asks for your approval first.

## Triage From Slack

When someone hits a problem, the useful moment is right then, in the channel where they said so. `/darcy triage` opens a form in Slack: describe what went wrong, optionally attach a video, and pick the TestDrive workspace it belongs to.

Darcy investigates against the real application and replies **in the thread** with what it found and the evidence behind it. If the problem is reproducible, it drafts a **triage candidate** — a proposed acceptance test capturing exactly what broke.

{% hint style="info" %}
**A triage candidate is quarantined until you say otherwise.** It's held apart from your suite and won't run with it. From the Slack thread, the reporter saves it — which promotes it into a real test — or discards it. Nothing enters your suite because someone reported a bug.
{% endhint %}

Triage needs the Slack workspace connected for your organization, TestDrive enabled, and access to the workspace you pick. Attached video is capped at 10 MB. See [Slack](/build/integration-overview/enterprise/slack.md) for connecting the workspace, and note you can run the same triage conversation from the workspace's own Darcy Chat panel.

## MCP: TestDrive in Your Editor

**MCP Install** in the workspace header gives you a ready-made configuration for an MCP client — your editor, or another agent. Once connected, the agent you already code with can list tests, start runs, poll for results, save scripts, and read or write application knowledge.

The practical loop: your editor agent changes code, runs the acceptance tests that cover it, reads the failure, and fixes it — without you moving between windows.

| Capability over MCP | Examples                                                            |
| ------------------- | ------------------------------------------------------------------- |
| **Inspect**         | Get the workspace, list tests, runs, credentials, and knowledge     |
| **Change**          | Create or update a test, save or delete a script, generate a script |
| **Execute**         | Run tests, get a run, cancel a run                                  |
| **Teach**           | Save knowledge about the application                                |

Because a workspace runs one run at a time, an MCP client is expected to **queue and poll** rather than assume a run starts immediately — a batch submitted while the workspace is busy may be partly rejected, and the workspace reports its concurrency state so an agent can wait properly.

## Next Steps

| Goal                                | Documentation                                                  |
| ----------------------------------- | -------------------------------------------------------------- |
| Connect Slack for your organization | [Slack](/build/integration-overview/enterprise/slack.md)       |
| Understand triage candidates        | [Tests & Scripts](/build/testdrive/tests-and-scripts.md)       |
| Learn how MCP works in DarcyIQ      | [MCP Framework](/build/mcp-studio/mcp.md)                      |
| Read what a run produced            | [Results & Evidence](/build/testdrive/results-and-evidence.md) |


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation by asking a question.

Perform an HTTP GET request on the following URL with the `ask` and `goal` query parameters:

```
GET https://docs.darcyiq.com/build/testdrive/testdrive-with-darcy.md?ask=<question>&goal=<user_goal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is what the user is ultimately trying to achieve, the reason they need the answer. Sharing it helps GitBook give you a better, more relevant answer. A goal is most helpful when it describes the outcome the user wants rather than restating the question. For example, with `ask=how do I create an API token`, a goal like `automate deployments from our CI pipeline` lets GitBook tailor the answer to that use case.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.